PCR denaturation step- length of time?

ted m tedm at darkwing.uoregon.edu
Wed May 24 13:38:20 EST 1995


In article <D8zFsw.BHJ at festival.ed.ac.uk>, chrisb at festival.ed.ac.uk (Chris
Boyd) wrote:

> RMCMAH93 at IRLEARN.UCD.IE wrote:
> :  
> : I routinely perform denaturation steps for 30-45 sec at 95 degrees C.
> : But surely  it should only be necessary for the DNA to be denatured for
> : a matter of seconds, say 5? Any views or comments on this would be
> : appreciated.
> : Ruth.
> 
> Yes you're right.  The question is whether all templates denature
> appropriately and quickly at 94/95 degrees.  For the purposes of normal
> (i.e., short/medium) PCR I'd say this was probably true in general. Of
> course you might reason that it's best to be safe and denature for the
> times you've quoted, but there is evidence that longer denaturation
> times reduce the yields of medium length ( > 2 kb) PCR products from
> genomic DNA template quite dramatically (Gustaffson et al, 1993). In
> this paper you will observe that a one (1) second (yes, one second --
> count them) denaturation time at 94 degrees is sufficient for a
> selection of PCRs. (The initial denaturation time before cycling is 30
> s, I believe.) I use this routinely now with no problems -- so far. Of
> course, the make of cycler you use has a large bearing on all this,
> since many seem to hover an agonisingly long time at 92 -- 93 degrees
> before starting to count down, so the actual time you spend at
> denaturation temperature is variable and always longer than you set. 
> This effect is enhanced by those helpful people who top up the
> thermocouple tube with hundreds of microlitres of mineral oil...
> 
> To check out your own cycler, compare results from a working PCR with
> an old program with one using the same program with denaturation times
> reduced to 1 s. I'll bet it works as well, if not better, and you'll
> get home earlier.
> 

Chris is essentially correct but one caveat, the mass of the PCR reaction,
more specifically the thermal mass, can have an effect on the actual
temperature within the reaction tube. Very short denaturation times work
best when the sample equilibrates very quickly. Thus the faster machines
all use thin wall tubes and relatively small volumes. A one sec
denaturation may not work on a thick walled 0.6ml tube with a 200ul
reaction and 100ul mineral oil, the actual temperature within the tube may
be a gradient from 96°->?° within the tube. The denaturation would be much
less efficient and could really show in genomic amplifications (low
yields, perhaps). It remains, of course to DO THE EXPERIMENT and see for
yourself, but it can work and work well. Another  advantage is that
reactions can be substantially more specific if the denaturation times
(and anneal times) are shorter.
                                                   Regards, Ted Michelini



More information about the Methods mailing list